No serial on ATiny841 2.0.0 Codeline

[ahshah]

Description: I can flash updates via the IDE using optiboot, however I can't get anything to print to the serial monitor.

Bootloader UART Settings: TX: PA1 RX: PA2 Board settings:

Heres the code:

// the setup function runs once when you press reset or power the board
void setup() {
  // initialize digital pin LED_PIN as an output.
  Serial.begin(115200);
}

// the loop function runs over and over again forever
void loop() {
  Serial.println("on");
  delay(1000);  // wait for a second
  Serial.println("off");
  delay(1000);  // wait for a second
}

Here's the hook up, please note vcc and gnd connections are below the breadboard adapting board. The FTDI is a sparkfun model.

[ahshah]

Codeline is currently on d670622155a459bf6e0757ac2032d0495bfa71b0: Correct issues with avrdude,conf.m make wiring readable

[ahshah]

The power and gnd rails read 3.3 v , and you can see 3.3v is soldered in the on the sparkfun FTDI controller

[SpenceKonde]

Curious indeed. There isn't much to go wrong in serial on 841 - but some changes were made so it is possible that something was broken, this is a very serious bug seeing as the whole reason this core was started was to get a tiny with 2 serial ports.

Can you post the .lst and .map files that are created when you export compiled binary? I know it does listings, and I think maps are in too, the map is actually what I'm more interested in for once...

[ahshah]

Hi please find the requested resources attached. Codeline remaints at https://github.com/SpenceKonde/ATTinyCore/commit/d670622155a459bf6e0757ac2032d0495bfa71b0 Serial.Attiny841.lst.map.zip

[SpenceKonde]

I may have fixed this issue ( am in middle of a move, don't have access to hardware to test on atm)

[ahshah]

Confirmed, thanks for the fix! Confirmed working with 747d8533b747758be3bc4e91eb5c35a05057467c. Could you narrow down what in the last few changes addressed this issue? Just for my edification.

[SpenceKonde]

It was that big commit with all those updates to the part-specific docs. That one also has a change to HardwareSerial (maybe several of the hardware serial files). It was failing to retain the register addresses. They were passed to the constructor and flushed down the toilet.

[SpenceKonde]

Thanks for testing!

[ahshah]

Hi, me again. Seems like I'm getting corruption on the serial when running at slower speeds. Similar sketch, different result:

// the setup function runs once when you press reset or power the board
void setup() {
  Serial.begin(9600);
}

// the loop function runs over and over again forever
void loop() {
  Serial.println("on");
  delay(1000);  // wait for a second
  Serial.println("off");
  delay(1000);  // wait for a second
}

Results in gibberish-y output like so:

offonofCo                                     offon
         ofC
oofCoofCo   ofCon
oVH      HonoffConoff
   oHooffConCoffC
oHoCHonCofCoofCoC5VHonofCoCofVon
                  o
oHo                ofVon
   5
ofCooofCon
oVHoCofVHon5
            onC

Serial settings is 9600 @ 8N1 I've tested and reproduced this corruption on the following two commit points, but not inbetween.

747d853 5508f88

Board settings as follows:

[ahshah]

Hi it looks like this issue with baudrate has to do with internal oscillator variance, as similarly discussed here: https://forum.arduino.cc/t/attiny-841-serial-comms-baud-rate/343701

My attiny 841 is running off 5v so it seems that this is the likely culprit